Ultrasound image-based thyroid nodule assessment system

Abstract

The application relates to the technical field of ultrasonic image analysis, and discloses a thyroid nodule evaluation system based on ultrasonic images. The system obtains medical record data and medical examination data of a patient through a multidimensional acquisition module, classifies and forms a data set, analyzes the thyroid function state of each patient through an intelligent evaluation module, generates a function coefficient, analyzes the picture quality of each group of ultrasonic images again, generates corresponding detection data group Jcsj, establishes a unified evaluation standard, quickly screens out key frames from a large amount of ultrasonic image data, has high multidimensional integration efficiency, sets a fixed range threshold value for the intelligent evaluation module, analyzes and generates a nodule volume and a nodule score in combination with a medical data set, and in combination with the function coefficient, the detection data group and the nodule score, evaluates the risk grade of the thyroid nodule and the picture quality of the ultrasonic image of the patient, screens out key frames of the ultrasonic image, and outputs a treatment suggestion; and the intelligent evaluation has high precision.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of ultrasonic image analysis, in particular to a thyroid nodule evaluation system based on ultrasonic images. BACKGROUND

[0002] Thyroid nodule refers to a localized mass formed by abnormal proliferation of thyroid cells, which is a common thyroid disease. Abnormal iodine intake is one of the common factors. When iodine is deficient, thyroid hormone synthesis is insufficient, and the thyroid will compensate for the proliferation to form nodules, and excessive iodine can also interfere with the normal function of the thyroid gland, inducing nodules. In addition, long-term mental stress, chemical radiation living environment, are all possible causes of thyroid nodules. The symptoms of thyroid nodules vary, most patients have no obvious symptoms, and are often found during physical examination, some patients may have a neck mass that moves up and down with swallowing, and sometimes may also be accompanied by pain, hoarseness and other symptoms. If the nodule is large, it may compress the surrounding tissue, causing difficulty breathing, difficulty swallowing and other problems. Thyroid nodules are divided into benign and malignant. Benign nodules grow slowly, have clear boundaries, and are soft in texture, while malignant nodules grow rapidly, have irregular boundaries, and are hard in texture. Once thyroid nodules are found, medical treatment is needed, and the nature of the nodules is determined through ultrasonic examination, CT examination, fine needle aspiration biopsy and other means. For benign nodules, regular reexamination or drug treatment can be taken according to the situation, and if it is a malignant nodule, surgical treatment is needed, and comprehensive treatment measures such as radiotherapy and chemotherapy are needed to ensure the health of the patient.

[0003] At present, the traditional ultrasonic image-based thyroid nodule evaluation system has insufficient ability to distinguish between cystic nodules and solid nodules, is prone to overdiagnosis, missed diagnosis, and misdiagnosis of complex nodules, has low treatment efficiency, lacks effective data integration mechanism and unified evaluation standard, and limits the accuracy of comprehensive diagnosis. SUMMARY

[0004] (I) Technical problems solved

[0005] In view of the defects of the prior art, the present application provides an ultrasonic image-based thyroid nodule evaluation system, which has the advantages of high multi-dimensional integration efficiency, intelligent evaluation precision, etc., and solves the problems of insufficient differentiation ability of the traditional ultrasonic image-based thyroid nodule evaluation system and low accuracy of comprehensive diagnosis.

[0006] (II) Technical solutions

[0007] To achieve the above-mentioned purpose, the present application provides the following technical solutions: an ultrasonic image-based thyroid nodule evaluation system, comprising a multi-dimensional acquisition module and an intelligent evaluation module.

[0008] The multi-dimensional acquisition module is composed of a patient data unit and an examination data unit, the patient data unit acquires a patient data set by connecting an electronic medical record system through a network, the patient data set includes medical record data of all patients, and the examination data unit acquires a medical data set by connecting a color Doppler ultrasound diagnostic instrument and a CT scanner through a network, the medical data set includes medical examination data of all patients;

[0009] The intelligent evaluation module is composed of a function analysis unit, an image analysis unit and an evaluation management unit, the function analysis unit analyzes the thyroid function state of each patient according to the patient data set, and generates a corresponding function coefficient Gnx, the image analysis unit analyzes the picture quality of each group of ultrasound images according to the medical data set, and generates a corresponding detection data group Jcsj, the image analysis unit is provided with a fixed range of length threshold CDY, volume threshold TJY and intensity threshold QDY, and further analyzes and generates nodule volume TJ and nodule score Jep in combination with the medical data set, the evaluation management unit is provided with a fixed range of function threshold GNY, score threshold JEPY, rate ratio threshold LBY, range threshold JCY and contrast threshold DBY, and further evaluates the risk level of thyroid nodules and the picture quality of ultrasound images of the patient in combination with the function coefficient Gnx, the detection data group Jcsj and the nodule score Jep, and outputs treatment suggestions.

[0010] Preferably, the expression of the patient data set is {H1 l , H2 l , H3 l ,..., Hz l}, H1 l to Hz l represent the medical record data of the first patient to the zth patient, the medical record data includes patient name, height, weight, high-risk environmental exposure frequency, TSH index, FT3 index and FT4 index, and l represents patient age.

[0011] Preferably, the expression of the medical data set is {Y1 s , Y2 s , Y3 s ,..., Yx s}, Y1 s to Yx s represent the first group to the xth group of medical examination data, the medical examination data includes patient name, ultrasound frequency, spatial resolution, pixel value, frame rate, gray scale contrast, number of nodules, nodule transverse diameter, nodule longitudinal diameter, nodule shape, nodule edge, nodule echo intensity and nodule blood supply grade, and s represents a specific time point of collecting medical examination data.

[0012] Preferably, the function coefficient Gnx calculation process is as follows:

[0013] According to the patient data set, the medical record data of the i th patient is extracted, and the height of the i th patient is marked as sg i The weight of the i th patient is marked as tz i The high-risk environmental exposure frequency of the i th patient is marked as bl i The TSH index of the i th patient is marked as TSH i The FT3 index of the i th patient is marked as FT3 i The FT4 index of the i th patient is marked as FT3 i The age of the i th patient is marked as i l ;

[0014]

[0015] In the formula, BMI i represents the body mass index of the i th patient;

[0016]

[0017] In the formula, α1 represents the weight of the patient's body mass index, α2 represents the weight of the patient's age and high-risk environmental exposure frequency ratio, α3 represents the weight of the patient's TSH index, α4 represents the weight of the patient's FT3 index, and α5 represents the weight of the patient's FT4 index., α1, α2, α3, α4 and α5 are constants, and α1+α2+α3+α4+α5=1, represents the functional coefficient Gnx of the i th patient calculated according to the weights α1, α2, α3, α4 and α5 i .

[0018] Preferably, the detection data set Jcsj calculation process is as follows:

[0019] According to the medical data set, the medical examination data of the i th patient is extracted, and the ultrasonic frequency of the i th patient during single ultrasonic examination is marked as bp, the spatial resolution of the i th patient's single ultrasonic image is marked as kf, the pixel value of the i th patient's single ultrasonic image is marked as xs, the frame rate of the i th patient during single ultrasonic examination is marked as zl, and the gray scale contrast of the i th patient's single ultrasonic image is marked as hd;

[0020]

[0021] In the formula, represents the ratio of ultrasonic frequency to ultrasonic image spatial resolution, xs max and xs min are the maximum and minimum values of the pixel value of the i th patient's single ultrasonic image, respectivelymax xs min denotes the range of the pixel value of the single ultrasound image of the i-th patient, denotes the ratio of the frame rate to the gray scale contrast of the ultrasound image, Jcsj i denotes the detection data set corresponding to the single ultrasound examination of the i-th patient Jcsj i .

[0022] Preferably, the nodule volume TJ calculation process is as follows:

[0023] According to the medical data set, the medical examination data of the i-th patient is extracted, and the transverse diameter of the nodule detected by CT examination of the i-th patient is marked as hj i The longitudinal diameter of the nodule detected by CT examination of the i-th patient is marked as zj i ;

[0024] TJ i = hj i × zj i

[0025] In the formula, TJ i denotes the nodule volume of the i-th patient.

[0026] 8. Preferably, the nodule score Jep analysis process is as follows:

[0027] If the number of nodules of the i-th patient is greater than 1, the nodule score Jep i of the i-th patient is added by 1;

[0028] If the transverse diameter hj i of the nodule of the i-th patient exceeds the length threshold CDY or the longitudinal diameter zj i of the nodule of the i-th patient exceeds the longitudinal diameter threshold ZJY, the nodule score Jep i of the i-th patient is added by 1;

[0029] If the nodule volume TJ i of the i-th patient exceeds the volume threshold TJY, the nodule score Jep i of the i-th patient is added by 1;

[0030] If the nodule morphology of the i-th patient is irregular, the nodule score Jep i of the i-th patient is added by 1;

[0031] If the nodule edge of the i-th patient is fuzzy, the nodule score Jep i of the i-th patient is added by 1;

[0032] If the nodule echo intensity of the i-th patient is lower than the intensity threshold QDY, the nodule score Jep i of the i-th patient is added by 1;

[0033] If the echo intensity of the nodule of the i-th patient exceeds the intensity threshold QDY, the nodule score Jep of the i-th patient is added by 2 i add 2;

[0034] If the morphology of the nodule of the i-th patient is irregular and the edge of the nodule is blurred, the nodule score Jep of the i-th patient is added by 3 i add 3;

[0035] If the blood supply grade of the nodule of the i-th patient is higher than grade I, the nodule score Jep of the i-th patient is added by 2. i add 2.

[0036] Preferably, when the function coefficient Gnx exceeds the function threshold GNY, it indicates that the thyroid function state of the patient is abnormal, and the corresponding risk level is a low level, and the patient is suggested to perform an ultrasound examination in time.

[0037] Preferably, when the number of the nodules is greater than 1 and the nodule score Jep is included in the score threshold JEPY, it indicates that the corresponding risk level of the thyroid nodule of the patient is a medium level, and the evaluation management unit screens the key frame of the ultrasound image according to the detection data group Jcsj, and the screening process is as follows:

[0038] If the ratio of the ultrasound frequency to the spatial resolution of the ultrasound image in the detection data group Jcsj is included in the rate ratio threshold LBY, the pixel value range of the ultrasound image is included in the range threshold JCY, and the ratio of the frame rate to the gray scale contrast of the ultrasound image is included in the contrast threshold DBY, it indicates that the picture quality of the ultrasound image is good, that is, the key frame.

[0039] Preferably, when the nodule score Jep exceeds the score threshold JEPY, it indicates that the corresponding risk level of the thyroid nodule of the patient is a high level, and it is suggested to perform a fine needle biopsy examination in time.

[0040] Compared with the prior art, the present application provides a thyroid nodule evaluation system based on ultrasound images, which has the following beneficial effects:

[0041] 1、The multi-dimensional acquisition module is connected with the electronic medical record system, the color Doppler ultrasound diagnostic instrument and the CT scanner in the application, the medical record data and the medical examination data of all patients are obtained, and the patient data set and the medical data set are classified and composed, the intelligent evaluation module analyzes the thyroid function state of each patient according to the patient data set, generates the corresponding function coefficient Gnx, comprehensively analyzes the different physiological indexes of the patient, and fully understands the health status of the patient, the intelligent evaluation module analyzes the picture quality of each group of ultrasound images according to the examination data set, generates the corresponding detection data group Jcsj, establishes a unified evaluation standard, quickly screens out the key frame from a large amount of ultrasound image data, and has high multi-dimensional integration efficiency.

[0042] 2、The application sets a fixed range of length threshold CDY, volume threshold TJY and intensity threshold QDY through the intelligent evaluation module, and analyzes and generates nodule volume TJ and nodule score Jep in combination with a medical data set, and the higher the final score is, the more the patient needs immediate treatment, the intelligent evaluation module is provided with a fixed range of function threshold GNY, score threshold JEPY, rate ratio threshold LBY, range threshold JCY and contrast threshold DBY, and in combination with function coefficient Gnx, detection data set Jcsj and nodule score Jep, the risk level of the patient's thyroid nodule and the picture quality of the ultrasound image are evaluated, the key frames of the ultrasound image are screened and treatment suggestions are output, and the intelligent evaluation has high accuracy. BRIEF DESCRIPTION OF DRAWINGS

[0043] Figure 1 It is a schematic diagram of the system structure of the application. DETAILED DESCRIPTION

[0044] The technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are only part of the embodiments of the application, not all the embodiments. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the application.

[0045] Since the conventional thyroid nodule evaluation system based on ultrasound images has insufficient ability to distinguish between cystic nodules and solid nodules, overdiagnosis, missed diagnosis and misjudgment of complex nodules are prone to occur, the treatment efficiency is low, there is a lack of effective data integration mechanism and unified evaluation standard, and the accuracy of comprehensive diagnosis is limited, therefore, a thyroid nodule evaluation system based on ultrasound images is provided, please refer to Figure 1 , the thyroid nodule evaluation system based on ultrasound images comprises a multi-dimensional acquisition module and an intelligent evaluation module.

[0046] The multi-dimensional acquisition module is composed of a patient data unit and an examination data unit, the patient data unit acquires a patient data set through a network connection electronic medical record system, the patient data set comprises medical record data of all patients, and the expression of the patient data set is {H1 l , H2 l , H3 l ,..., Hz l}, H1 l to Hz lThe medical record data of the first to the zth patient, the medical record data including the patient's name, height, weight, high-risk environmental exposure frequency, TSH index, FT3 index and FT4 index, l represents the patient's age, long-term exposure to radiation, chemicals and other harmful environments also increases the risk of thyroid nodules, thyroid stimulating hormone TSH can stimulate the thyroid to synthesize and release free triiodothyronine FT3 and free thyroxine FT4, by monitoring the level changes of TSH, FT3 and FT4, the functional status of the thyroid can be more accurately evaluated;

[0047] The examination data unit collects medical data sets through network connection of color Doppler ultrasound diagnostic instrument and CT scanner, the medical data set including medical examination data of all patients, the expression of the medical data set is {Y1 s 、Y2 s 、Y3 s 、...、Yx s}, Y1 s to Yx s represent the first to the xth group of medical examination data, the medical examination data including the patient's name, ultrasound frequency, spatial resolution, pixel value, frame rate, gray scale contrast, number of nodules, nodule transverse diameter, nodule longitudinal diameter, nodule morphology, nodule edge, nodule echo intensity and nodule blood supply grade, specifically, the color Doppler ultrasound diagnostic instrument provides the ultrasound frequency, spatial resolution, pixel value, frame rate, gray scale contrast, nodule morphology, nodule edge, nodule echo intensity and nodule blood supply grade parameters, and the CT scanner provides the number of nodules, nodule transverse diameter and nodule longitudinal diameter parameters, in the actual evaluation process, the nodule echo intensity includes low intensity, medium intensity and high intensity, the low intensity indicates that the nodule contains more liquid or cystic components inside, the medium intensity indicates that the nodule internal organization is uniform, and the high intensity indicates that the nodule contains more substantial components inside, the nodule blood supply grade is divided into levels I-III, the level I indicates that the nodule has little or no blood supply inside, which is usually the characteristic of benign nodules, the level II indicates that the nodule has less blood supply inside, which has a certain probability of being benign or malignant, and the level III indicates that the nodule has very rich blood supply inside, which has a large number of blood vessels supply, which is usually the characteristic of malignant nodules, s represents the specific time point of collecting medical examination data, and the multi-dimensional data is helpful for subsequent analysis and evaluation of the nature of thyroid nodules and development of the corresponding treatment plan;

[0048] The intelligent evaluation module is composed of a function analysis unit, an image analysis unit and an evaluation management unit, the function analysis unit analyzes the thyroid function status of each patient according to the patient data set, and generates the corresponding function coefficient Gnx, and the calculation process is as follows:

[0049] According to the patient data set, the medical record data of the ith patient is extracted, and the height of the ith patient is marked as sg i , and the weight of the ith patient is marked as tzi Let the high-risk environment exposure frequency of the i-th patient be denoted as bl i Let the TSH index of the i-th patient be denoted as TSH i Let the FT3 index of the i-th patient be denoted as FT3 i Let the FT4 index of the i-th patient be denoted as FT3 i Let the age of the i-th patient be denoted as i l ;

[0050]

[0051] In the formula, BMI i represents the body mass index of the i-th patient;

[0052]

[0053] In the formula, α1 represents the weight for the body mass index of the patient, α2 represents the weight for the ratio of the patient's age and high-risk environment exposure frequency, α3 represents the weight for the TSH index of the patient, α4 represents the weight for the FT3 index of the patient, α5 represents the weight for the FT4 index of the patient, α1, α2, α3, α4 and α5 are all constants, and α1+α2+α3+α4+α5=1, represents the functional coefficient Gnx of the i-th patient calculated according to the weights α1, α2, α3, α4 and α5 i Comprehensive analysis of different physiological indicators of the patient helps to fully understand the health status of the patient;

[0054] The image analysis unit analyzes the picture quality of each group of ultrasound images according to the ultrasound data set, and generates the corresponding detection data group Jcsj, and the calculation process is as follows:

[0055] According to the examination data set, the medical examination data of the i-th patient is extracted, and the ultrasound frequency of the i-th patient during single ultrasound examination is denoted as bp, the spatial resolution of the i-th patient single ultrasound image is denoted as kf, the pixel value of the i-th patient single ultrasound image is denoted as xs, the frame rate of the i-th patient during single ultrasound examination is denoted as zl, and the gray scale contrast of the i-th patient single ultrasound image is denoted as hd;

[0056]

[0057] In the formula, represents the ratio of the ultrasound frequency to the spatial resolution of the ultrasound image, xs max and xs min are the maximum and minimum values of the pixel value of the i-th patient single ultrasound image, respectively, and xs max -xmin a range of the i-th patient's single ultrasound image pixel value, a ratio of frame rate and ultrasound image gray scale contrast, Jcsj i a detection data set corresponding to the i-th patient's single ultrasound examination Jcsj i , establish a unified evaluation standard, quickly screen out key frames in a large amount of ultrasound image data, and multi-dimensionally integrate high efficiency;

[0058] The image analysis unit is provided with a fixed range of length threshold CDY, volume threshold TJY and intensity threshold QDY, and further combines the ultrasound data set to analyze and generate nodule volume TJ and nodule score Jep;

[0059] The nodule volume TJ calculation process is as follows:

[0060] According to the examination data set, the medical examination data of the i-th patient is extracted, and the transverse diameter of the nodule checked by CT of the i-th patient is marked as hj i The longitudinal diameter of the nodule checked by CT of the i-th patient is marked as zj i ;

[0061] TJ i = hj i × zj i

[0062] In the formula, TJ i represents the nodule volume of the i-th patient;

[0063] The nodule score Jep analysis process is as follows:

[0064] If the number of nodules of the i-th patient is greater than 1, the nodule score Jep i of the i-th patient is added by 1;

[0065] If the transverse diameter hj i of the nodule of the i-th patient exceeds the length threshold CDY or the longitudinal diameter zj i of the nodule of the i-th patient exceeds the longitudinal diameter threshold ZJY, the nodule score Jep i of the i-th patient is added by 1;

[0066] If the nodule volume TJ i of the i-th patient exceeds the volume threshold TJY, the nodule score Jep i of the i-th patient is added by 1;

[0067] If the nodule morphology of the i-th patient is irregular, the nodule score Jep i of the i-th patient is added by 1;

[0068] If the nodule edge of the i-th patient is fuzzy, the nodule score Jep iAdd 1;

[0069] If the echo intensity of the nodule of the i-th patient is lower than the intensity threshold QDY, it indicates that the nodule contains more liquid or cystic components, and the probability of cystic nodule is greater, and the nodule score Jep of the i-th patient i Add 1,

[0070] If the echo intensity of the nodule of the i-th patient exceeds the intensity threshold QDY, it indicates that the nodule contains more substantial components, and the probability of solid nodule is greater, and the nodule score Jep of the i-th patient i Add 2;

[0071] If the nodule of the i-th patient is irregular in shape and the edge of the nodule is blurred, the nodule score Jep of the i-th patient i Add 3;

[0072] If the blood supply grade of the nodule of the i-th patient is higher than grade I, the nodule score Jep of the i-th patient i Add 2, the higher the final score, the more the patient needs immediate treatment;

[0073] The evaluation management unit is provided with a fixed range of function threshold GNY, score threshold JEPY, rate ratio threshold LBY, range threshold JCY and contrast threshold DBY, combined with the function coefficient Gnx, the detection data set Jcsj and the nodule score Jep, the risk level of the patient's thyroid nodule and the picture quality of the ultrasound image are evaluated. When the function coefficient Gnx exceeds the function threshold GNY, it indicates that the patient's thyroid function is abnormal, and the corresponding risk level is low, and the patient is suggested to have an ultrasound examination in time. When the number of nodules is greater than 1 and the nodule score Jep is included in the score threshold JEPY, it indicates that the corresponding risk level of the patient's thyroid nodule is medium, and the evaluation management unit screens the key frames of the ultrasound image according to the detection data set Jcsj, and the screening process is as follows:

[0074] If the ratio of the ultrasound frequency to the spatial resolution of the ultrasound image in the detection data set Jcsj is included in the rate ratio threshold LBY, the pixel value range of the ultrasound image is included in the range threshold JCY, and the ratio of the frame rate to the gray scale contrast of the ultrasound image is included in the contrast threshold DBY, it indicates that the picture quality of the ultrasound image is good, that is, the key frame;

[0075] When the nodule score Jep exceeds the score threshold JEPY, it indicates that the corresponding risk level of the patient's thyroid nodule is high, and it is suggested to have a fine needle puncture examination in time. Smaller nodules may not need immediate treatment, while larger nodules may need surgical resection. The intelligent evaluation has high accuracy.

[0076] Example 1:

[0077] In this embodiment, a 35-year-old male youth with a height of 1.75m was selected as the experimental subject. The youth's weight was 70 kg, his high-risk environmental exposure frequency was 7 times / month, and his TSH level was 2 mIU / L, FT3 level was 3.5 pmol / L, and FT4 level was 12 pmol / L. The functional coefficient Gnx of this youth was calculated as follows:

[0078]

[0079] In the formula, the youth's Body Mass Index (BMI) i The value is 22.9. 0.2 represents the weight of the patient's body mass index, the weight of the ratio of the patient's age to the frequency of exposure to high-risk environments, the weight of the patient's TSH level, the weight of the patient's FT3 level, and the weight of the patient's FT4 level. α1, α2, α3, α4, and α5 are all constants, and 0.2 + 0.2 + 0.2 + 0.2 + 0.2 = 1. Based on the weights of α1, α2, α3, α4, and α5, the functional coefficient Gnx of this young person is obtained. i It is 9.08.

[0080] Example 2:

[0081] In this embodiment, a color Doppler ultrasound diagnostic instrument with a frequency of 2MHz and a frame rate of 60 frames / second was selected as the experimental object. The acquired ultrasound images had a spatial resolution of 0.5mm, a pixel value range of 50-255, and a grayscale contrast ratio of 12. The calculation process for the detection data set Jcsj of this set of ultrasound images is as follows:

[0082]

[0083] In the formula, 0.25 represents the ratio of ultrasound frequency to ultrasound image spatial resolution, 205 represents the range of pixel values ​​of a single ultrasound image for the i-th patient, and 0.2 represents the ratio of frame rate to ultrasound image grayscale contrast.

[0084] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. An ultrasound image-based thyroid nodule assessment system, characterized by: It comprises a multi-dimensional acquisition module and an intelligent evaluation module. The multi-dimensional acquisition module is composed of a patient data unit and an examination data unit, the patient data unit acquires a patient data set by connecting an electronic medical record system through a network, the patient data set comprises medical record data of all patients, and the examination data unit acquires a medical data set by connecting a color Doppler ultrasound diagnostic instrument and a CT scanner through a network, the medical data set comprises medical examination data of all patients. The intelligent evaluation module is composed of a function analysis unit, an image analysis unit and an evaluation management unit, the function analysis unit analyzes the thyroid function state of each patient according to the patient data set, and generates a corresponding function coefficient Gnx, and the image analysis unit analyzes the picture quality of each group of ultrasound images according to the medical data set, and generates a corresponding detection data group Jcsj. The detection data group Jcsj calculation process is as follows: According to the medical data set, the medical examination data of the ith patient is extracted, the ultrasound frequency of the single ultrasound examination of the ith patient is marked as bp, the spatial resolution of the single group of ultrasound images of the ith patient is marked as kf, the pixel value of the single group of ultrasound images of the ith patient is marked as xs, the frame rate of the single ultrasound examination of the ith patient is marked as zl, and the gray scale contrast of the single group of ultrasound images of the ith patient is marked as hd. In the formula, represents the ratio of the ultrasonic frequency to the spatial resolution of the ultrasonic image, xs max and xs min respectively represent the maximum value and the minimum value of the pixel value of the single group of ultrasonic images of the i th patient, xs max -xs min represents the range of the pixel value of the single group of ultrasonic images of the i th patient, represents the ratio of the frame rate to the gray scale contrast of the ultrasonic image, Jcsj i represents the detection data group corresponding to the single ultrasonic examination of the i th patient, Jcsj i ; The image analysis unit is provided with a fixed range of length threshold CDY, volume threshold TJY and intensity threshold QDY, and further combined with the medical data set, a nodule volume TJ and a nodule score Jep are generated, the evaluation management unit is provided with a fixed range of function threshold GNY, score threshold JEPY, rate ratio threshold LBY, range threshold JCY and contrast threshold DBY, and further combined with the function coefficient Gnx, the detection data group Jcsj and the nodule score Jep, the risk level of the patient's thyroid nodule and the picture quality of the ultrasound image are evaluated, the key frames of the ultrasound image are screened and the treatment suggestion is output.

2. The ultrasound image-based thyroid nodule assessment system of claim 1, wherein: The expression of the patient data set is {H1 l , H2 l , H3 l , …, Hz l}, H1 l to Hz l represents the medical record data of the first to the zth patient, and the medical record data includes patient name, height, weight, high-risk environmental exposure frequency, TSH index, FT3 index and FT4 index, and l represents patient age.

3. The ultrasound image-based thyroid nodule assessment system of claim 2, wherein: The expression of the medical data set is {Y1 s , Y2 s , Y3 s , …, Yx s}, Y1 s to Yx s represent the first group to the xth group of medical examination data, and the medical examination data includes patient name, ultrasonic frequency, spatial resolution, pixel value, frame rate, gray scale contrast, nodule number, nodule transverse diameter, nodule longitudinal diameter, nodule morphology, nodule edge, nodule echo intensity, and nodule blood supply grade, and s represents a specific time point at which the medical examination data is collected.

4. The ultrasound image-based thyroid nodule assessment system of claim 3, wherein: The function coefficient Gnx calculation process is as follows: According to the patient data set, the medical record data of the ith patient is extracted, and the height of the ith patient is marked as sg i The weight of the ith patient is marked as tz i The high-risk environmental exposure frequency of the ith patient is marked as bl i The TSH index of the ith patient is marked as TSH i The FT3 index of the ith patient is marked as FT3 i The FT4 index of the ith patient is marked as FT3 i The age of the ith patient is marked as i l ; In the formula, BMI i represents the body mass index of the i-th patient; In the formula, a1 represents a weight for a body mass index of the patient, a2 represents a weight for a ratio of an age of the patient and a high-risk environment exposure frequency, a3 represents a weight for a TSH index of the patient, a4 represents a weight for a FT3 index of the patient, a5 represents a weight for a FT4 index of the patient, a1, a2, a3, a4 and a5 are all constants, and a1+a2+a3+a4+a5=1, Gnx represents a functional coefficient of the i-th patient calculated according to the a1, a2, a3, a4 and a5 weights i .

5. The ultrasound image-based thyroid nodule assessment system of claim 4, wherein: The nodule volume TJ calculation process is as follows: According to the medical data set, the medical examination data of the i-th patient is extracted, and the transverse diameter of the nodule checked by CT of the i-th patient is marked as hj i The longitudinal diameter of the nodule checked by CT of the i-th patient is marked as zj i ; TJ i = hj i × zj i In the formula, TJ i denotes the volume of the nodule of the i-th patient.

6. The ultrasound image-based thyroid nodule assessment system of claim 5, wherein: The nodule score Jep analysis process is as follows: Jep= the number of nodules in the i-th patient, if the number of nodules in the i-th patient is greater than 1 i + 1; if the i-th patient's nodule transverse diameter hj i if the i-th patient's nodule longitudinal diameter zj exceeds the length threshold CDY i the i-th patient's nodule score Jep when the i-th patient's nodule longitudinal diameter zj exceeds the longitudinal diameter threshold ZJY i add 1; If the i-th patient's nodule volume TJ i If the i-th patient's nodule volume TJ exceeds the volume threshold TJY, the i-th patient's nodule score Jep i Add 1; Jep= 1 if the i-th patient has irregular nodule morphology, otherwise Jep= 0 i Add 1; Jep= 1 if the i-th patient has a fuzzy nodule edge i add 1; if the echogenicity of the nodule of the i-th patient is lower than the intensity threshold QDY, the nodule score Jepof the i-th patient is set to 1 i add 1; If the echogenicity of the nodule of the i-th patient exceeds the intensity threshold QDY, the nodule score Jepof the i-th patient is incremented by 2. i Add 2; If the i-th patient has irregular nodular morphology and the i-th patient has a nodular margin that is fuzzy, the i-th patient's nodular score Jep i Add 3; If the blood supply of the nodule of the i-th patient is higher than grade I, the nodule score Jepof the i-th patient is i Add 2.

7. The ultrasound image-based thyroid nodule assessment system of claim 6, wherein: When the function coefficient Gnx exceeds the function threshold GNY, it indicates that the thyroid function state of the patient is abnormal, and the corresponding risk level is low, and the patient is suggested to have an ultrasound examination in time.

8. The ultrasound image-based thyroid nodule assessment system of claim 7, wherein: When the number of nodules is greater than 1 and the nodule score Jep is included in the score threshold JEPY, it indicates that the corresponding risk level of the patient's thyroid nodule is medium, and the evaluation management unit screens the key frames of the ultrasound image according to the detection data group Jcsj, and the screening process is as follows: When the ratio of the ultrasound frequency to the ultrasound image spatial resolution in the detection data group Jcsj is included in the rate ratio threshold LBY, the ultrasound image pixel value range is included in the range threshold JCY, and the ratio of the frame rate to the ultrasound image gray scale contrast is included in the contrast threshold DBY, it indicates that the picture quality of the ultrasound image is good, that is, the key frame.

9. The ultrasound image-based thyroid nodule assessment system of claim 8, wherein: When the nodule score Jep exceeds the score threshold JEPY, it indicates that the corresponding risk level of the patient's thyroid nodule is high, and fine needle aspiration examination is suggested in time.

Images